Scaffold-mediated non-viral delivery of nucleic acids for sustained gene silencing in regenerative medicine

• Main Author: Chin, Jiah Shin
• Other Authors: Chew Sing Yian
• Format: Thesis-Doctor of Philosophy
• Language: English
• Published: Nanyang Technological University 2022
• Subjects: Engineering::Materials; Science::Medicine
• Online Access: https://hdl.handle.net/10356/158706
• Institution: Nanyang Technological University

Treatments with nucleic acid therapeutics have emerged as a promising approach since it addresses the molecular causes of hindered tissue repair by manipulating gene expression profiles in targeted cells within the injured tissue system. Although nucleic acid-based therapy has seen significant advancement in various tissue repair applications, its success has largely been hindered by the mode of delivery. Highly efficient viral gene vectors and carriers are typically used for effective transfection to occur. However, significant safety issues and complications have also arisen out of such viral delivery strategies. Non-viral nucleic acid delivery strategies offer improved safety profiles and are promising alternatives but unfortunately, the limited transfection efficiencies of non-viral delivery platforms must first be addressed before achieving functional tissue repair outcomes. Moreover, existing non-viral delivery of nuclei acids largely rely on hydrogels and particles. None of which mimics the injury environment nor can they sustain gene knockdown. Accordingly, the aim of this thesis was to design and fabricate electrospun fibers that mimic the injury environment while incorporating small nucleic acids that silence genes upregulated upon injury. Specifically, such synergistic delivery of topographical and biochemical cues aims to enhance tissue repair in skin and spinal cord. While both tissue environments are vastly different, the key barrier that prevents healing in chronic wounds and after a spinal cord injury (SCI) is the chronic and sustained high levels of inflammation. Silencing genes upregulated during inflammation provides an alternative to mitigating the damage in chronic wounds and after SCI. Wet electrospinning and air gap electrospinning were the two different techniques used to fabricate poly(ε-caprolactone) (PCL), rat-tail type 1 collagen and poly (caprolactone-co-ethyl ethylene phosphate) (PCLEEP) scaffolds that mimicked the native extracellular matrix in the skin and axons, respectively. Wet electrospun PCL scaffolds promoted cell migration, vascularization and re-epithelialisation with minimal inflammatory response. These scaffolds were then incorporated with Connexin 43 (Cx43) - specific antisense oligodeoxynucleotides (asODNs). However, due to the limitations of the wet electrospinning technique and the choice of polymer, the mass of Cx43 asODNs delivered to wound bed was insufficient to downregulate Cx43 expression at the epithelial tongue. Hence, no significant difference in epithelial tongue thickness or migration distance was observed. These findings were consolidated and then extended into a more complex SCI environment where a fiber-hydrogel scaffold for sustained delivery of Cx43 asODN, while providing synergistic topographical cues to guide axonal ingrowth was fabricated. Here, as an extension of my previous work on delivering nucleic acids to SCI injury environment, PCLEEP, instead of PCL, was used to accelerate degradation. Materials used in wounds could ideally be rapidly integrated or pushed out along with the scab as the wound heals. In contrast, accessibility to scaffolds implanted at SCI sites is low and hence, scaffolds cannot be easily removed. Thus, the need to be bioresorbable is more apparent in SCI applications. These scaffolds demonstrated the sustained release of Cx43 asODN for up to 25 days. In addition, Cx43 up-regulation after complete transection SCI in rats was supressed, preserving neurons around the injury site, promoting axonal extension while decreasing glial scarring and microglial activation after SCI. Beyond non-viral delivery of nucleic acids, CRISPR/Cas9 components were also delivered in a localized and non-viral manner via electrospun scaffolds. Specifically, using mussel-inspired bioadhesive coating, polyDOPA-melanin (pDOPA), Cas9: sgRNA lipofectamine complexes were adsorbed onto bio-mimicking fiber scaffolds. As this is a proof of concept, PCL was chosen as the choice of polymer to be electrospun especially since no specific application nor injury environment was decided. U2OS.EGFP cells took up Cas9: sgRNA lipofectamine complexes directly from the scaffolds via reverse transfection and expression of EGFP in these cells was successfully knockdown. In vitro studies use cells derived from animals or cell lines which have an infinite lifespan. While these model systems are relatively cheap and simple to purchase, they fail to capture the inherent complexity of organ systems. The use of animals in in vivo studies addresses many of the shortcomings of in vitro studies. Hence, the bulk of this thesis relies on animal models to evaluate nucleic acids loaded scaffolds. However, the problem of translatability remains. Given the considerable physiological differences between human and animals, the animal models used for evaluation might not truly reflect the functional outcome when translated to humans. In this thesis, a novel perturbed wound model that more accurately recapitulates features of human chronic wounds for more accurate testing of biofunctionalized scaffolds was established. Numerous features such as hyper-thickened epithelial tongue, delayed wound closure, chronic inflammatory environment, overexpression of Cx43, presence of senescent cells and ECM degradation at wound edges were observed in the perturbed wounds. These features accurately recapitulated the elements that hinder chronic wound healing in humans. Given its relevance to chronic wounds, this perturbed wound model potentially serves as a more relevant platform for testing of wound healing therapeutics. Altogether, this thesis demonstrated the feasibility of electrospinning to fabricate scaffolds that mimic vastly different injury environments while non-virally delivering gene-silencing nucleic acids for regenerative medicine. Importantly, the methodology can be extended to more complicated gene editing techniques (e.g. CRISPR). Additionally, a new perturbed model that more accurately recapitulates features of human chronic wounds was established in hopes of serving as a more relevant platform for testing of wound healing therapeutics and biofunctionalized scaffolds.